Tag
This article details how a systematic fund replaced its traditional NLP pipeline with a RAG-based LLM agent architecture, achieving a 340% improvement in alpha generation from unstructured data. It cites recent research (Alpha-GPT 2.0, FinCon, FinAgent) showing significant gains in automated factor discovery and trading performance.
This research tests whether Benjamin Graham's classical value investing rules can act as a mathematical 'low-pass filter' to prevent modern machine learning models (XGBoost, AutoGluon) from overfitting to market noise. Using 20 years of S&P 500 data, the authors find that Graham's rules combined with Random Forest achieve high returns with lower risk than complex AI models alone.
A comprehensive guide explaining the Kalman filter and its application in building smarter trading systems, including mathematical foundations and production-grade examples.
A thread introducing Loop Engineering as a solution to the common problem of quant strategies that backtest perfectly but fail in live trading, emphasizing the need for iterative optimization.
This paper introduces CARLOS, a deep reinforcement learning algorithm that learns continuous-time optimal stopping rules for American-style options using an aggregate deep neural network, effectively closing the Bermudan-American value gap with high computational efficiency.
Recommend 11 high-quality open-source projects on GitHub covering AI agent frameworks, AI programming, memory systems, research automation, and quantitative investment tools, designed to help developers get started quickly and boost efficiency.
A Chinese quantitative finance tutorial for absolute beginners, using Jupyter Notebook format, containing 4 chapters (quantitative cognition, return analysis, dual moving average strategy, and strategy backtesting). Uses yfinance to fetch real data, each chapter can be run through in about 30 minutes.
PandaAI proposes a closed-loop neuro-symbolic LLM agent for sequential decision-making in quantitative finance, integrating market regime modeling and constrained alpha generation to address low SNR and non-stationarity in financial data, achieving significant improvements over state-of-the-art time-series models.
A programmer earning $385,000 per year failed his interview at Jane Street for refusing to use Claude Code, reflecting that AI tools have become industry entry standards. On Polymarket, there are bets on the penetration speed of AI tools.
This tweet introduces three financial/engineering tools open-sourced by top quantitative institutions such as Jane Street, Goldman Sachs, and J.P. Morgan: magic-trace (high-precision process tracing), gs-quant (Python package for derivatives pricing and risk management), and Perspective (real-time data visualization tool), helping quant enthusiasts gain institutional-level technical capabilities for free.
This article introduces a 50-minute video shared by a Jane Street quant trader on YouTube, covering a full roadmap for quantitative learning, including practical tips on Polymarket, which is a valuable reference for quant practitioners.
Kronos is the world's first open-source foundational large model for financial markets, trained from scratch on 12 billion real candlestick data points, supporting price prediction and volatility forecasting, far outperforming general models, and completely free and open-source.
The article promotes a Stanford lecture on Markov Decision Processes as a valuable resource for understanding the mathematical foundations of systematic trading, claiming it offers more insight than a short-term internship at major financial firms.
An MIT professor who trains quants for top hedge funds delivered a closed-door keynote at Oxford for Man Group, and the 1-hour recording was accidentally left on a public server. This free resource offers valuable insights into advanced quantitative finance and analytical methodologies.
This paper introduces Semantic State Abstraction Interfaces (SSAI) to separate representation hypotheses from optimization variance in LLM-augmented portfolio decisions. It concludes that SSAI's apparent advantage is largely a basket-selection effect, with dense encodings and principal components performing better empirically.
A researcher claims to have achieved an 83% return on real markets using Neural Networks and Hidden Markov Models, publishing both the theory and an implementation guide for Polymarket.
A 29-year-old Oklahoma sales consultant claims to have built an Ethereum price prediction system using Claude and multiple AI agents, replacing an entire quant team and allegedly generating over $300,000 in monthly profits. The content originates from social media, its authenticity is questionable, and it carries clear signs of marketing promotion.
NautilusTrader is an open-source, Rust-native algorithmic trading engine for multi-asset, multi-venue systems, providing a single event-driven architecture for research, simulation, and live execution with Python or Rust strategy development.
Kronos is an open-source foundation model for financial K-line sequences, trained on data from over 45 global exchanges. It uses a specialized tokenizer and a decoder-only Transformer, and has been accepted at AAAI 2026.